Samuel Felton

Centre National de la Recherche Scientifique

Papers

Total Citations

H-Index

About

Samuel Felton is a pioneering researcher at the intersection of robotics and machine learning, whose work is redefining how robots perceive and interact with their environments. His primary research areas center on visual servoing—a technique that uses camera feedback to control robot motion—and the innovative application of deep learning to this field. Felton’s major contributions include developing methods that operate in autoencoder latent spaces, effectively merging the precision of photometric approaches like Direct Visual Servoing with the robust convergence of feature-based methods. His most-cited paper, “Visual Servoing in Autoencoder Latent Space” (2022, 14 citations), introduced a novel framework for extracting visual information directly from compressed image representations, bypassing traditional hand-crafted features. Building on this, his 2023 work, “Deep metric learning for visual servoing: when pose and image meet in latent space” (9 citations), further advanced the field by learning a metric space that unifies pose and image data, enabling more accurate and stable robot control. Though early in his career, Felton’s work is gaining traction for its potential to simplify and enhance robotic manipulation in unstructured settings, marking him as a rising star in embodied AI research.